Hi Rechard,
I tried "zgexpv", but failed. Below is the script:

from expokit import *
import numpy as np
from scipy.sparse import random
from scipy.sparse.linalg import onenormest,expm_multiply,aslinearoperator
n=1000
m=100
v = np.random.ranf(n)
v /= np.linalg.norm(v)
w = np.empty(n)
A = random(n,n,format="csr")
A = -1j*(A.T + A)/2
anorm = onenormest(A)
wsp = np.zeros(7+n*(m+2)+5*(m+2)*(m+2),dtype=np.float64)
iwsp = np.zeros(m+2,dtype=np.int32)
matvec = aslinearoperator(A)
zgexpv(n, m,1.0, v, w, 0.0, anorm,wsp, len(wsp), iwsp, len(iwsp), matvec, 0, 0)

I got the following error:
Traceback (most recent call last):
File "test1.py", line 16, in
zgexpv(n, m,1.0, v, w, 0.0, anorm,wsp, len(wsp), iwsp, len(iwsp), matvec, 0, 0)
expokit.error: failed in processing argument list for call-back matvec.
Call-back argument must be function|instance|instance.call|f2py-function but got int.

Can you provide an example for the use of zgexpv?